3 replies to this topic

[marchese85]

i'm working on dynamic simulation of batch distillation in hysys!!!
in my system there are three components
1 ethylacetate
2 water
3 isopropanol
during the simulation there are two liquid phase in the reflux accumulator (the components at 102 kpa and 70 °C determines liquid-liquid equilibrium)
i would use a different set of binary coefficient (NRTL model) for diffrent equipments. The accumulator, reboiler and other equipments are in different environment from the column.
how can i fix a set of binary coefficient for the only accumulator (coefficient from the liquid -liquid equilibrium data regression) and another for the reboiler , column and remaining equipments (from the liquid vapor equilibrium data regression )?????

i attached a file in which there is my simulation!!!!!!
thanks in advance
best regards
p.s. sorry for my english

[Chellani]

Hi,
You can use a different FP with your new BIPs (estimated using LLE). Here are the steps (hope it makes sense)
- Add a new Fluid Package (FP)
- Use NRTL as your property package (PP)
- Use Binary Interaction Parameters (BIPs) / binary coefficients estimated using LLE for this FP
- Add a stream cutter in the downstream of ‘E-100’
- Use newly defined FP as the outlet FP of this stream cutter and select any transition basis for this cutter
- Attach product stream of this cutter to accumulator
- Similarly use a cutter in the downstream of accumulator. It’s not mandatory as HYSYS would automatically perform FP propagation through column environment

[MrShorty]

how can i fix a set of binary coefficient for the only accumulator (coefficient from the liquid -liquid equilibrium data regression) and another for the reboiler , column and remaining equipments (from the liquid vapor equilibrium data regression )?????

This is an interesting question, because there really isn't yet (using the NRTL equation anyway) a good answer to it. If I may quote from the 2nd ed of [i]Molecular Thermodynamics of Fluid-Phase Equilibria" by Prausnitz, Lichtenthaler, and de Azevedo:

In mixtures with miscisibility gaps, it is difficult to find an expression for the molar excess Gibbs energy gE which yields simultaneously correct activity coefficients in the miscible region and correct activity coefficients at the solubility limits. ...it is not possible to describe accurately both vapor-liquid and liquid-liquid equilibria using a simple expression for gE containing only two adjustable binary parameters.

My own experience bears this out. Any time I want to use a single set of parameters in the NRTL equation to represent both VLE and LLE data, it is usually some kind of "compromise" that is chosen based on the individual circumstance. If you want to fix a single set of NRTL parameters, you're going to probably need to do the same thing: find a set of "compromise" parameters and see if they are accurate enough to use.

[milenkom]

i'm working on dynamic simulation of batch distillation in hysys!!!
in my system there are three components
1 ethylacetate
2 water
3 isopropanol
during the simulation there are two liquid phase in the reflux accumulator (the components at 102 kpa and 70 °C determines liquid-liquid equilibrium)
i would use a different set of binary coefficient (NRTL model) for diffrent equipments. The accumulator, reboiler and other equipments are in different environment from the column.
how can i fix a set of binary coefficient for the only accumulator (coefficient from the liquid -liquid equilibrium data regression) and another for the reboiler , column and remaining equipments (from the liquid vapor equilibrium data regression )?????

i attached a file in which there is my simulation!!!!!!
thanks in advance
best regards
p.s. sorry for my english

This one shouldn't be to difficult. I used to do steady state simulation of azeotropic distillation and for instance in Aspen Plus this is straightforward. You create new thermodynamic property and assign this package only to the unit of interest. Now if it's possible in Aspen It can be done in Hysys as well. I'm pretty sure I've been doing this in Hysys as well, but that was already some time ago and I can't remember how.